A high proportion of the women in both groups had felt confident during pregnancy in their capacity to breastfeed. Perhaps this explains, why the mothers with diabetes despite the daily diabetes related pressures, did breastfeed to such a high extent at both 2 and 6 months after childbirth. A high degree of self-efficacy has been found elsewhere to positively influence women’s decision to breastfeed. Another encouraging result in our study is that the overall experience of breastfeeding was positive for almost all women in both groups, and especially for the mothers with diabetes. We have previously reported that these women experience a demanding time during the first few months with their newborn. It would be reasonable to believe that other aggravating circumstances, such as the high frequency of caesarean sections, maternal complications etc., among women with diabetes can have an impact on health outcomes postpartum. However, these variables were not shown to explain the shorter duration of breastfeeding but are likely to explain why initiation and establishment were more delayed in the diabetic group.
At both 2 and 6 months after childbirth, mothers with diabetes reported less need to organize their time than before compared to the reference mothers. This may indicate that the mothers with diabetes already had distinct daily structures in place. While everyday life does not have to be highly organized and structured for most people, it often is for mothers with diabetes. As has been described previously[15, 22, 23, 25], for them organizing and predicting everyday life means balancing activities with managing blood glucose levels. This daily challenge might be, at the same time, the explanation for the somewhat inconsistent result that although the mothers with diabetes reported less need to organize their time, they remained more affected by disruptions in daily life at 6 months postpartum than the reference group mothers. Being accustomed to structure as mothers managing their diabetes may have made them more sensitive to changes and disruptions even before they entered motherhood and began breastfeeding. This suggested explanation has to be scientifically proven in future studies.
The study also focused on women’s concerns about their own health, and we found that the mothers with diabetes reported being more worried about their health during the entire study period compared to the reference group. The reason for this was not studied. One explanation might be that their need for insulin, which had increased 2 to 3 times during pregnancy, suddenly decreased afterwards and was frequently less than it was before pregnancy. Hypoglycemic episodes in relation to breastfeeding have also been reported to be more frequent, sometimes even dramatic episodes were experienced. Another explanation might be that the increased daily interruptions related to care for the newborn or increased sensitivity to disruptions affected the mothers’ level of concern. This suggested explanation also needs to be further explored.
Another study of a general population of women in Sweden (aged 38 and 50) showed that perceived high frequency of interruptions was related to poor subjective health among the younger women, and to low satisfaction with life as a whole in both groups. Women with children living at home and/or who lived with a partner experienced disturbing interruptions more frequently than those without children living at home or those who lived alone. Because frequent interruptions and changes in daily activities may constitute a risk for a reduction in subjective health, it might be beneficial to advise women with diabetes to minimize the risk for interruptions and changes in routines after childbirth. Our previous findings showed a negative effect on psychological well-being the more breastfeeding affected diabetes management. Since one of the most common sources of interruptions and daily minor hassles in women’s daily life are family members, it might be important to inform those living with new mothers with diabetes about their increased sensitivity to interruptions in daily life and an extended need for support to reduce such daily minor hassles to reduce risk of stress and ill health. Previous studies have shown that peer supporters are more likely than health care professionals to provide an authentic presence related to sharing experiences. Mothers with diabetes often lack access to supportive persons with common experiences of diabetes and childbearing[16, 28]. Sharing experiences about constructive ways of balancing everyday life and of managing unpredictable blood glucose levels while dealing with daily activities during breastfeeding and childcare might reduce their extraordinary challenges.
There are several limitations to be aware of in this study. The relatively small sample size might have implications for the power of the study as statistical power calculations were calculated on variables not central in this study. Another limitation is that due to logistical obstacles we have no information on the drop-out rate in the reference group.
The use of a questionnaire which has not been psychometrically evaluated is a further limitation. We did not find any psychometrically evaluated instrument covering the broad aspects of interest in this study. Unfortunately, the Breastfeeding Self-Efficacy Scale had not yet been translated and validated to a Swedish population. We did, however, test our questionnaire for face validity in both groups of women. Similar statements about daily structure and interruptions have been used in other studies with women[26, 30] which may thereby increase the applicability of these statements in the questionnaire. A factor that supports the use of a specifically designed questionnaire is the need to use contextual questions in specific and unique situations. Detailed validation might be needed in some areas. When exploring new areas, a creative and pragmatic approach has to be used, and there are many examples of useful data being derived from simple tools in different studies. Nevertheless, these preliminary findings require replication and further investigation in order to draw conclusions.
The strength of our study is the use of matched controls with same parity and childbirth in the same gestational week, as it allows comparison between two groups with a crucial difference such as living with or without a demanding chronic illness – diabetes, for example.